z/VM on Hercules

[herculessearch]

Can I run z/VM on top of a Hercules installation?

Thanks in advance for your help.

[Josef Sipek]

Post by herculessearch
Can I run z/VM on top of a Hercules installation?

Yes and no. As far as I know, there's no _technical_ problem, and hercules
can handle it just fine, but you need IBM to let you run z/VM - licensing
issues.

Josef "Jeff" Sipek.

[William D Carroll]

Like was stated, Technically it will run but proper licensing is required.

Problem is the installation, unless you install from TAPE or do a second
level install using the DVDPRIME you can not install it

The DVD/CD needs the SYSG console which is not supported by Hercules.



Doug



From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of Josef Sipek
Sent: Thursday, June 21, 2007 5:32 PM
To: hercules-390-***@public.gmane.org
Subject: Re: [hercules-390] z/VM on Hercules

Post by herculessearch
Can I run z/VM on top of a Hercules installation?

Yes and no. As far as I know, there's no _technical_ problem, and hercules
can handle it just fine, but you need IBM to let you run z/VM - licensing
issues.

Josef "Jeff" Sipek.
--
The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all progress
depends on the unreasonable man.
- George Bernard Shaw





[Non-text portions of this message have been removed]

[Johnny Luo]

William,

I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.

I didn't see anything specifal. Do I miss something here?

Post by William D Carroll
Like was stated, Technically it will run but proper licensing is required.
Problem is the installation, unless you install from TAPE or do a second
level install using the DVDPRIME you can not install it
The DVD/CD needs the SYSG console which is not supported by Hercules.
Doug
Behalf Of Josef Sipek
Sent: Thursday, June 21, 2007 5:32 PM
Subject: Re: [hercules-390] z/VM on Hercules

Post by herculessearch
Can I run z/VM on top of a Hercules installation?

Yes and no. As far as I know, there's no _technical_ problem, and hercules
can handle it just fine, but you need IBM to let you run z/VM - licensing
issues.
Josef "Jeff" Sipek.
--
The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all progress
depends on the unreasonable man.
- George Bernard Shaw
[Non-text portions of this message have been removed]

[William D Carroll]

That's a Demo Pak + zVM 5.3 which only ESP site's have at this time (until
june 29th anyways)

Hercules lacks the SYSG support needed by zVM upto 5.2 (I can not speak for
5.3 except to say if it was designed for Flex-es they may have changed it to
allow an install from a normal console) which is an OSA-ICC Connected
Console Device.











From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of Johnny Luo
Sent: Friday, June 22, 2007 8:46 AM
To: hercules-390-***@public.gmane.org
Subject: Re: [hercules-390] z/VM on Hercules



William,

I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.

I didn't see anything specifal. Do I miss something here?

On 6/22/07, William D Carroll <voxr-gmqx98DRti9Wk0Htik3J/***@public.gmane.org
<mailto:voxr%40smgvbest.com> > wrote:



.


<http://geo.yahoo.com/serv?s=97359714/grpId=342064/grpspId=1707281942/msgId=
50957/stime=1182516454/nc1=4507179/nc2=3848642/nc3=4670553>




[Non-text portions of this message have been removed]

[Rich Smrcina]

z/VM doesn't require SYSG support. Just a non-SNA 3270 console device,
if it happens to be SYSG that's fine.

Post by William D Carroll
That's a Demo Pak + zVM 5.3 which only ESP site's have at this time (until
june 29th anyways)
Hercules lacks the SYSG support needed by zVM upto 5.2 (I can not speak for
5.3 except to say if it was designed for Flex-es they may have changed it to
allow an install from a normal console) which is an OSA-ICC Connected
Console Device.
Behalf Of Johnny Luo
Sent: Friday, June 22, 2007 8:46 AM
Subject: Re: [hercules-390] z/VM on Hercules
William,
I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.
I didn't see anything specifal. Do I miss something here?
.
<http://geo.yahoo.com/serv?s=97359714/grpId=342064/grpspId=1707281942/msgId=
50957/stime=1182516454/nc1=4507179/nc2=3848642/nc3=4670553>
[Non-text portions of this message have been removed]
http://groups.yahoo.com/group/hercules-390
http://www.hercules-390.org
Yahoo! Groups Links

--
Rich Smrcina
VM Assist, Inc.
Phone: [phone]
Ans Service: 360-715-2467
rich.smrcina at vmassist.com
http://www.linkedin.com/in/richsmrcina

Catch the WAVV! http://www.wavv.org
WAVV 2008 - Chattanooga - April 18-22, 2008

[William D Carroll]

It requires SYSG support for the INSTALLATION of z/VM from DVD

It does not require SYSG support once installed.







From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of Rich Smrcina
Sent: Friday, June 22, 2007 9:47 AM
To: hercules-390-***@public.gmane.org
Subject: Re: [hercules-390] z/VM on Hercules



z/VM doesn't require SYSG support. Just a non-SNA 3270 console device,
if it happens to be SYSG that's fine.

Post by William D Carroll
That's a Demo Pak + zVM 5.3 which only ESP site's have at this time (until
june 29th anyways)
Hercules lacks the SYSG support needed by zVM upto 5.2 (I can not speak for
5.3 except to say if it was designed for Flex-es they may have changed it to
allow an install from a normal console) which is an OSA-ICC Connected
Console Device.

[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
] On

Post by William D Carroll
Behalf Of Johnny Luo
Sent: Friday, June 22, 2007 8:46 AM
Subject: Re: [hercules-390] z/VM on Hercules
William,
I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.
I didn't see anything specifal. Do I miss something here?

<mailto:voxr%40smgvbest.com>

Post by William D Carroll
.

<http://geo.yahoo.com/serv?s=97359714/grpId=342064/grpspId=1707281942/msgId=

Post by William D Carroll
50957/stime=1182516454/nc1=4507179/nc2=3848642/nc3=4670553>
[Non-text portions of this message have been removed]

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>

Post by William D Carroll
http://groups.yahoo.com/group/hercules-390
http://www.hercules-390.org
Yahoo! Groups Links

--
Rich Smrcina
VM Assist, Inc.
Phone: [phone]
Ans Service: 360-715-2467
rich.smrcina at vmassist.com
http://www.linkedin.com/in/richsmrcina

Catch the WAVV! http://www.wavv.org
WAVV 2008 - Chattanooga - April 18-22, 2008





[Non-text portions of this message have been removed]

[Paul Raulerson]

Not required, but sure convenient and cost effective! J

-Paul





From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of William D Carroll
Sent: Friday, June 22, 2007 10:35 AM
To: hercules-390-***@public.gmane.org
Subject: RE: [hercules-390] z/VM on Hercules



It requires SYSG support for the INSTALLATION of z/VM from DVD

It does not require SYSG support once installed.

From: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
] On
Behalf Of Rich Smrcina
Sent: Friday, June 22, 2007 9:47 AM
To: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
Subject: Re: [hercules-390] z/VM on Hercules

z/VM doesn't require SYSG support. Just a non-SNA 3270 console device,
if it happens to be SYSG that's fine.

Post by William D Carroll
That's a Demo Pak + zVM 5.3 which only ESP site's have at this time (until
june 29th anyways)
Hercules lacks the SYSG support needed by zVM upto 5.2 (I can not speak for
5.3 except to say if it was designed for Flex-es they may have changed it to
allow an install from a normal console) which is an OSA-ICC Connected
Console Device.

<mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
] On

Post by William D Carroll
Behalf Of Johnny Luo
Sent: Friday, June 22, 2007 8:46 AM

<mailto:hercules-390%40yahoogroups.com>

Post by William D Carroll
Subject: Re: [hercules-390] z/VM on Hercules
William,
I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.
I didn't see anything specifal. Do I miss something here?

<mailto:voxr%40smgvbest.com>
<mailto:voxr%40smgvbest.com>

Post by William D Carroll
.

<http://geo.yahoo.com/serv?s=97359714/grpId=342064/grpspId=1707281942/msgId=

Post by William D Carroll
50957/stime=1182516454/nc1=4507179/nc2=3848642/nc3=4670553>
[Non-text portions of this message have been removed]

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>

Post by William D Carroll
http://groups.yahoo.com/group/hercules-390
http://www.hercules-390.org
Yahoo! Groups Links

--
Rich Smrcina
VM Assist, Inc.
Phone: [phone]
Ans Service: 360-715-2467
rich.smrcina at vmassist.com
http://www.linkedin.com/in/richsmrcina

Catch the WAVV! http://www.wavv.org
WAVV 2008 - Chattanooga - April 18-22, 2008

[Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]

[William D Carroll]

If you can install z/VM from the DVD without having an existing VM
installation so you can use the DVDPRIME EXEC

I'd sure like to know how you get past the hardcoded SYSG console device in
DVD ?



I am not talking a second level install, I've been through this many times
on this list.

To install directly from DVD, first level requires SYSG support, you can
not use a CONS= statement for first level on

zVM =< 5.2, I don't know about 5.3 yet.



So if you know a way to install first level from the DVD please let me and
others know?



Thanks

Doug





From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of Paul Raulerson
Sent: Saturday, June 23, 2007 8:24 AM
To: hercules-390-***@public.gmane.org
Subject: RE: [hercules-390] z/VM on Hercules



Not required, but sure convenient and cost effective! J

-Paul

From: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
] On
Behalf Of William D Carroll
Sent: Friday, June 22, 2007 10:35 AM
To: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
Subject: RE: [hercules-390] z/VM on Hercules

It requires SYSG support for the INSTALLATION of z/VM from DVD

It does not require SYSG support once installed.

From: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
] On
Behalf Of Rich Smrcina
Sent: Friday, June 22, 2007 9:47 AM
To: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
Subject: Re: [hercules-390] z/VM on Hercules

z/VM doesn't require SYSG support. Just a non-SNA 3270 console device,
if it happens to be SYSG that's fine.

Post by William D Carroll
That's a Demo Pak + zVM 5.3 which only ESP site's have at this time (until
june 29th anyways)
Hercules lacks the SYSG support needed by zVM upto 5.2 (I can not speak for
5.3 except to say if it was designed for Flex-es they may have changed it to
allow an install from a normal console) which is an OSA-ICC Connected
Console Device.

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
] On

Post by William D Carroll
Behalf Of Johnny Luo
Sent: Friday, June 22, 2007 8:46 AM

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>

Post by William D Carroll
Subject: Re: [hercules-390] z/VM on Hercules
William,
I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.
I didn't see anything specifal. Do I miss something here?

<mailto:voxr%40smgvbest.com>
<mailto:voxr%40smgvbest.com>
<mailto:voxr%40smgvbest.com>

Post by William D Carroll
.

<http://geo.yahoo.com/serv?s=97359714/grpId=342064/grpspId=1707281942/msgId=

Post by William D Carroll
50957/stime=1182516454/nc1=4507179/nc2=3848642/nc3=4670553>
[Non-text portions of this message have been removed]

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>

Post by William D Carroll
http://groups.yahoo.com/group/hercules-390
http://www.hercules-390.org
Yahoo! Groups Links

--
Rich Smrcina
VM Assist, Inc.
Phone: [phone]
Ans Service: 360-715-2467
rich.smrcina at vmassist.com
http://www.linkedin.com/in/richsmrcina

Catch the WAVV! http://www.wavv.org
WAVV 2008 - Chattanooga - April 18-22, 2008

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]

[Paul Raulerson]

I was referring to: "It does not require SYSG support once installed."



You can also get around it via installing from tape images of course. It
sure seems like a lot of work though. It would be better indeed if IBM would
either just legitimize Herc someway or another. The open source licenses
appears to be a negative issue on this perhaps.



Anyway, yes, if you install from DVD, you need SYSG. How much trouble would
it be to emulate SYSG? And is there any other reason to do so given how
Hercules operates?



-Paul



From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of William D Carroll
Sent: Saturday, June 23, 2007 1:00 PM
To: hercules-390-***@public.gmane.org
Subject: RE: [hercules-390] z/VM on Hercules



If you can install z/VM from the DVD without having an existing VM
installation so you can use the DVDPRIME EXEC

I'd sure like to know how you get past the hardcoded SYSG console device in
DVD ?

I am not talking a second level install, I've been through this many times
on this list.

To install directly from DVD, first level requires SYSG support, you can
not use a CONS= statement for first level on

zVM =< 5.2, I don't know about 5.3 yet.

So if you know a way to install first level from the DVD please let me and
others know?

Thanks

Doug

From: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
] On
Behalf Of Paul Raulerson
Sent: Saturday, June 23, 2007 8:24 AM
To: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
Subject: RE: [hercules-390] z/VM on Hercules

Not required, but sure convenient and cost effective! J

-Paul

From: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
] On
Behalf Of William D Carroll
Sent: Friday, June 22, 2007 10:35 AM
To: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
Subject: RE: [hercules-390] z/VM on Hercules

It requires SYSG support for the INSTALLATION of z/VM from DVD

It does not require SYSG support once installed.

From: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
] On
Behalf Of Rich Smrcina
Sent: Friday, June 22, 2007 9:47 AM
To: hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
Subject: Re: [hercules-390] z/VM on Hercules

z/VM doesn't require SYSG support. Just a non-SNA 3270 console device,
if it happens to be SYSG that's fine.

Post by William D Carroll
That's a Demo Pak + zVM 5.3 which only ESP site's have at this time (until
june 29th anyways)
Hercules lacks the SYSG support needed by zVM upto 5.2 (I can not speak for
5.3 except to say if it was designed for Flex-es they may have changed it to
allow an install from a normal console) which is an OSA-ICC Connected
Console Device.

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
[mailto:hercules-390-***@public.gmane.org <mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
] On

Post by William D Carroll
Behalf Of Johnny Luo
Sent: Friday, June 22, 2007 8:46 AM

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>

Post by William D Carroll
Subject: Re: [hercules-390] z/VM on Hercules
William,
I once successfully installed z/vm 5.3 Demopkg DVDs for Flex-es under
hercules.
I didn't see anything specifal. Do I miss something here?

<mailto:voxr%40smgvbest.com>
<mailto:voxr%40smgvbest.com>
<mailto:voxr%40smgvbest.com>
<mailto:voxr%40smgvbest.com>

Post by William D Carroll
.

<http://geo.yahoo.com/serv?s=97359714/grpId=342064/grpspId=1707281942/msgId=

Post by William D Carroll
50957/stime=1182516454/nc1=4507179/nc2=3848642/nc3=4670553>
[Non-text portions of this message have been removed]

<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-subscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-unsubscribe%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>
<mailto:hercules-390-owner%40yahoogroups.com>

Post by William D Carroll
http://groups.yahoo.com/group/hercules-390
http://www.hercules-390.org
Yahoo! Groups Links

--
Rich Smrcina
VM Assist, Inc.
Phone: [phone]
Ans Service: 360-715-2467
rich.smrcina at vmassist.com
http://www.linkedin.com/in/richsmrcina

Catch the WAVV! http://www.wavv.org
WAVV 2008 - Chattanooga - April 18-22, 2008

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]

[Jay Maynard]

Post by Paul Raulerson
You can also get around it via installing from tape images of course. It
sure seems like a lot of work though. It would be better indeed if IBM would
either just legitimize Herc someway or another. The open source licenses
appears to be a negative issue on this perhaps.

How do you mean?

Post by Paul Raulerson
Anyway, yes, if you install from DVD, you need SYSG. How much trouble would
it be to emulate SYSG? And is there any other reason to do so given how
Hercules operates?

I don't know if anyone has any idea how hard it would be to emulate. I don't
believe it's documented.

Is there a reason to emulate it? Sure! Hercules is intended to be as
complete an emulation of the IBM system as we can make. That includes every
peripheral we can figure out.

[Mike Schwab]

A SYSG console is a standard 3270 emulator, but how do you designate
what connection is the SYSG console?

Post by Jay Maynard

Post by Paul Raulerson
Anyway, yes, if you install from DVD, you need SYSG. How much trouble would
it be to emulate SYSG? And is there any other reason to do so given how
Hercules operates?

I don't know if anyone has any idea how hard it would be to emulate. I don't
believe it's documented.
Is there a reason to emulate it? Sure! Hercules is intended to be as
complete an emulation of the IBM system as we can make. That includes every
peripheral we can figure out.
--

[Jay Maynard]

Post by Mike Schwab
A SYSG console is a standard 3270 emulator, but how do you designate
what connection is the SYSG console?

It's a 3270 emulator, but how is it presented to software running on the
mainframe?

Designating the connection to be used for the SYSG console would not be a
big deal; it could be selected in the same way that console groups or
addresses are selected now.

[Ivan Warren]

Post by Mike Schwab
A SYSG console is a standard 3270 emulator, but how do you designate
what connection is the SYSG console?

I don't think that's the issue.. I think the problem is figuring out how
the interface to SERVC (assuming SYSG is using SERVC the same way
SYSCONS is) is designed..

If that is sorted out, then having hercules relate a tn3270 session to
SYSG is just a matter of simple engineering.. (a special port, tn3270
terminal negotiation, you name it..)

--Ivan

[Phil Dickinson]

How about doing CTC support and IUCV support first so that we can make
better use of the freely available and always legal JES3 on MVS 3.8J in the
case of CTC, and MUSIC/SP in the case of IUCV.

:-)

Please.

Pretty please. With honey on it.

Post by Ivan Warren

Post by Mike Schwab
A SYSG console is a standard 3270 emulator, but how do you designate
what connection is the SYSG console?

I don't think that's the issue.. I think the problem is figuring out how
the interface to SERVC (assuming SYSG is using SERVC the same way
SYSCONS is) is designed..
If that is sorted out, then having hercules relate a tn3270 session to
SYSG is just a matter of simple engineering.. (a special port, tn3270
terminal negotiation, you name it..)
--Ivan

[Non-text portions of this message have been removed]

[Jay Maynard]

Post by Phil Dickinson
How about doing CTC support and IUCV support first so that we can make
better use of the freely available and always legal JES3 on MVS 3.8J in the
case of CTC, and MUSIC/SP in the case of IUCV.

I'd love to see both, myself...and have occasionally considered doing IUCV,
even though I don't know anything about it.

In the open source world, though, what gets added to the package is what
people sit down and do, and that's influenced by wha they want to see in the
package. It's called "scratching their own itch".

[Phil Dickinson]

I wish I could help, but I don't understand C, or any of the modern
languages and development packages... in fact I only ever just got by with
assembler.

Like a few others on the list I try to contribute when I can in the areas
that I know... a bit of SNA and a bit of RACF.

I'll just have to wait until someone that knows how to do this stuff wants
it badly enough.

Post by Jay Maynard

Post by Phil Dickinson
How about doing CTC support and IUCV support first so that we can make
better use of the freely available and always legal JES3 on MVS 3.8J in

the

Post by Phil Dickinson
case of CTC, and MUSIC/SP in the case of IUCV.

I'd love to see both, myself...and have occasionally considered doing IUCV,
even though I don't know anything about it.
In the open source world, though, what gets added to the package is what
people sit down and do, and that's influenced by wha they want to see in the
package. It's called "scratching their own itch".
--
Jay Maynard, K5ZC http://www.conmicro.com
http://jmaynard.livejournal.com http://www.tronguy.net
http://www.hercules-390.org (Yes, that's me!)
Buy Hercules stuff at http://www.cafepress.com/hercules-390

[Non-text portions of this message have been removed]

[Ivan Warren]

Post by Phil Dickinson
How about doing CTC support and IUCV support first so that we can make
better use of the freely available and always legal JES3 on MVS 3.8J in the
case of CTC, and MUSIC/SP in the case of IUCV.
:-)
Please.
Pretty please. With honey on it.

Ok.. The ins & outs..

Those 2 issues are on the priority list.. Only they are not 'simple'
stuff..

Actually, IUCV may be the easier part.. However, we really want to think
how we're going to be implementing IUCV in the first place.. (remember,
IUCV is only a transport service.. We want to keep our options open as
to how it is implemented so that we can use it for more than just IP
network interface or Socket emulation.. For example, the Linux/390 guys
have started implementing a AF_IUCV socket address family ! so we want
to be able to allow those folks to play with our implementation of IUCV !)

CTC is another issue.. the consensus (ok. maybe not a consensus, but how
I see things), to do this right, is we're going to need to redesign a
few things here and there.. And CTC will probably need a central entity
to deal with the CTC Finite State Machine that's need to implement a
general purpose CTC implementation.. So we need to sit on the drawing
board first !

But we'll get there, eventually..

SYSG, on the other hand, may just be a matter of a couple hundred lines
of C code, without a need to redesign anything or deep thinking (as soon
as we can figure out the host SYSG interface), so it may come first even
if it's not high priority (simply because it may be simpler to implement)..

And again, just like Jay mentioned : This is open source ! so anyone
with an elegant and architecturally correct design & implementation is
welcome to come forward.

I'm still not satisfied with at least ONE gory details in herc.. the I/O
subsystem has a design problem that's going to need to be addressed -
It's a major undertaking to fix a marginal[1] problem.. but it's an
architectural issue, so it's going to need to be addressed eventually !

--Ivan

[1] Marginal because it hasn't bothered anyone.. yet.. major because it
doesn't do what the POO says !

[Phil Dickinson]

Ivan,

Thanks for the well thought out response to my plea for CTC and IUCV.

As I said before, I am no developer, but if there is anything that I can do
to help this process along please let me know.

Phil.

Post by Ivan Warren
Ok.. The ins & outs..
Those 2 issues are on the priority list.. Only they are not 'simple'
stuff..

[Non-text portions of this message have been removed]

[Harold Grovesteen]

What is the architectural problem with the I/O subsystem, out of curiosity?

Harold Grovesteen

Post by Jay Maynard

Post by Phil Dickinson
How about doing CTC support and IUCV support first so that we can make
better use of the freely available and always legal JES3 on MVS 3.8J

in the

Post by Phil Dickinson
case of CTC, and MUSIC/SP in the case of IUCV.
:-)
Please.
Pretty please. With honey on it.

Ok.. The ins & outs..
Those 2 issues are on the priority list.. Only they are not 'simple'
stuff..
Actually, IUCV may be the easier part.. However, we really want to think
how we're going to be implementing IUCV in the first place.. (remember,
IUCV is only a transport service.. We want to keep our options open as
to how it is implemented so that we can use it for more than just IP
network interface or Socket emulation.. For example, the Linux/390 guys
have started implementing a AF_IUCV socket address family ! so we want
to be able to allow those folks to play with our implementation of IUCV !)
CTC is another issue.. the consensus (ok. maybe not a consensus, but how
I see things), to do this right, is we're going to need to redesign a
few things here and there.. And CTC will probably need a central entity
to deal with the CTC Finite State Machine that's need to implement a
general purpose CTC implementation.. So we need to sit on the drawing
board first !
But we'll get there, eventually..
SYSG, on the other hand, may just be a matter of a couple hundred lines
of C code, without a need to redesign anything or deep thinking (as soon
as we can figure out the host SYSG interface), so it may come first even
if it's not high priority (simply because it may be simpler to
implement)..
And again, just like Jay mentioned : This is open source ! so anyone
with an elegant and architecturally correct design & implementation is
welcome to come forward.
I'm still not satisfied with at least ONE gory details in herc.. the I/O
subsystem has a design problem that's going to need to be addressed -
It's a major undertaking to fix a marginal[1] problem.. but it's an
architectural issue, so it's going to need to be addressed eventually !
--Ivan
[1] Marginal because it hasn't bothered anyone.. yet.. major because it
doesn't do what the POO says !

[Ivan Warren]

Post by Harold Grovesteen
What is the architectural problem with the I/O subsystem, out of curiosity?

Well.. In the current state of affairs, for a write type operation, the
data is pre-fetched from storage for the length indicated in the CCW. In
the real world, data is fetched on demand by the Control Unit. This has
several implications (among others) :
- Immediate commands have to be pre-declared (an 'immediate command' is
a write type operation that returns a Channel End before any data is
consumed by the control unit)
- When some or all of the data is not accessible by the channel
subsystem, a write type operation to a not-ready device yields a channel
access exception instead of a Unit Check only
- When data is partially accessible, but the Control Unit would only
consume the accessible portion, a channel access exception is unduly
recognized. That may be the most troubling issue.
- When data is partially accessible, but the Control Unit would also
consume the non-accessible portion, a channel access exception is
recognized before any data is transfered to the Control Unit. This
should lead to partial writes, something that never occurs. This may be
a problem for Tape devices which then may lead to the tape being in the
wrong position..

For read type operations, there is an almost similar problem.. A read
type operation will always be taken to completion even if the storage
area addressed by the CCW cannot be written to by the Channel Subsystem.

There may be other conditions..

Note that in the 'real world', some similar conditions may arise,
especially when dealing with control units with buffering capabilities
(like DASD storage directors) when they are attached to Block
Multiplexor channels or when operating in burst mode. But even then, in
some cases, the 'real world' case would behave differently (think of a
large operations like Write Multiple Count Key And Data..)

Control Unit in hercules is the same as 'Device Handler' (the code
emulating a device and its control unit)
Channel access exception is when the channel subsystem tries to access
data that is either key storage protected, beyond addressing or lies
within an invalidated storage frame.

The problem with the current design is that it's not possible to have
the device handler 'ask' for data.. The entire data is always presented
as part of the CCW execution call (except when the CCW command code is
known to be an immediate command at which time no data is ever fetched)
and IDAW CCWs always have their data pre-assembled.

Ok.. So all of this is dealing with possibly marginal error cases..
(except for the "immediate command" for which we did find a workaround),
and no problem should hit on a well written I/O handling program. But
still.. I'm not entirely satisfied with the whole thing.

Clear as mud ?

--Ivan

[Binyamin Dissen]

On Sun, 24 Jun 2007 18:14:40 +0200 Ivan Warren <ivan-lnHwE90NT89Ooi3Kub+***@public.gmane.org> wrote:

:>Harold Grovesteen wrote:
:>> What is the architectural problem with the I/O subsystem, out of curiosity?

:>Well.. In the current state of affairs, for a write type operation, the
:>data is pre-fetched from storage for the length indicated in the CCW. In
:>the real world, data is fetched on demand by the Control Unit. This has
:>several implications (among others) :
:>- Immediate commands have to be pre-declared (an 'immediate command' is
:>a write type operation that returns a Channel End before any data is
:>consumed by the control unit)

Doesn't that depend on ORB & CCW bit combinations?

Or are just the CCW's pre-fetcheed?

[Ivan Warren]

Post by Binyamin Dissen
Doesn't that depend on ORB & CCW bit combinations?
Or are just the CCW's pre-fetcheed?

No.. That's just CCW prefetch, not the data ! (besides, in S/370 mode,
there is no such thing, and CCWs are not prefetched).

CCW prefetch (or lack thereof when indicated in the ORB or in S/370 mode
[1]) has to do with the possibility for self modifying CCWs or a program
modifying a hot CCW..

--Ivan

[1] Except under VM where CCWs are always prefetched - and that's why
there is a Diagnose to indicate to CP that the program has modified a
CCW while it was active, otherwise CP may not notice since CCWs are
prefetched and rewritten by CP.. (Note that you can only modify a TIC to
a NOP in that case !)

[Harold Grovesteen]

Totally clear, actually. I have seen some of how this works in the
code itself but had not taken that to the architectural implications.

Thanks,
Harold Grovesteen

Post by Harold Grovesteen

Post by Harold Grovesteen
What is the architectural problem with the I/O subsystem, out of

curiosity?
Well.. In the current state of affairs, for a write type operation, the
data is pre-fetched from storage for the length indicated in the CCW. In
the real world, data is fetched on demand by the Control Unit. This has
- Immediate commands have to be pre-declared (an 'immediate command' is
a write type operation that returns a Channel End before any data is
consumed by the control unit)
- When some or all of the data is not accessible by the channel
subsystem, a write type operation to a not-ready device yields a channel
access exception instead of a Unit Check only
- When data is partially accessible, but the Control Unit would only
consume the accessible portion, a channel access exception is unduly
recognized. That may be the most troubling issue.
- When data is partially accessible, but the Control Unit would also
consume the non-accessible portion, a channel access exception is
recognized before any data is transfered to the Control Unit. This
should lead to partial writes, something that never occurs. This may be
a problem for Tape devices which then may lead to the tape being in the
wrong position..
For read type operations, there is an almost similar problem.. A read
type operation will always be taken to completion even if the storage
area addressed by the CCW cannot be written to by the Channel Subsystem.
There may be other conditions..
Note that in the 'real world', some similar conditions may arise,
especially when dealing with control units with buffering capabilities
(like DASD storage directors) when they are attached to Block
Multiplexor channels or when operating in burst mode. But even then, in
some cases, the 'real world' case would behave differently (think of a
large operations like Write Multiple Count Key And Data..)
Control Unit in hercules is the same as 'Device Handler' (the code
emulating a device and its control unit)
Channel access exception is when the channel subsystem tries to access
data that is either key storage protected, beyond addressing or lies
within an invalidated storage frame.
The problem with the current design is that it's not possible to have
the device handler 'ask' for data.. The entire data is always presented
as part of the CCW execution call (except when the CCW command code is
known to be an immediate command at which time no data is ever fetched)
and IDAW CCWs always have their data pre-assembled.
Ok.. So all of this is dealing with possibly marginal error cases..
(except for the "immediate command" for which we did find a workaround),
and no problem should hit on a well written I/O handling program. But
still.. I'm not entirely satisfied with the whole thing.
Clear as mud ?
--Ivan

[Paul Raulerson]

From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of Jay Maynard
Sent: Saturday, June 23, 2007 6:55 PM
To: hercules-390-***@public.gmane.org
Subject: Re: [hercules-390] z/VM on Hercules

You can also get around it via installing from tape images of course. It
sure seems like a lot of work though. It would be better indeed if IBM would
either just legitimize Herc someway or another. The open source licenses
appears to be a negative issue on this perhaps.

How do you mean?

If it could be presented to IBM in way that allows the bean counters some
sense of control, illusionary as it may be.

Or another words, if IBM could put out a proprietary version of Hercules, as
an OCO release, they would probably be happier. <grin>

Anyway, yes, if you install from DVD, you need SYSG. How much trouble would
it be to emulate SYSG? And is there any other reason to do so given how
Hercules operates?

I don't know if anyone has any idea how hard it would be to emulate. I don't
believe it's documented.

Is there a reason to emulate it? Sure! Hercules is intended to be as
complete an emulation of the IBM system as we can make. That includes every
peripheral we can figure out.

It's part of the HMC, which would be very cool to emulate, but a lot of the
functions just don't really have counterparts in the Hercules world. ECP's
and so forth and so on. And SYSG can be assigned to the GUI interface, or
you can use the crippled copy of PCOM that serves as the 3270 console.

-Paul
--
Jay Maynard, K5ZC http://www.conmicro.com
http://jmaynard.livejournal.com http://www.tronguy.net
http://www.hercules-390.org (Yes, that's me!)
Buy Hercules stuff at http://www.cafepress.com/hercules-390





[Non-text portions of this message have been removed]

[ddsoftsystems]

It's possible install from DVD/CD as a VM second level.
After this you can use to as first level.
I used a z/VM 4.4 to install a z/VM 5.1 from DVD. But you need to use FBA disks and the correct device model, in this case is 9346.

Post by William D Carroll
Like was stated, Technically it will run but proper licensing is required.
Problem is the installation, unless you install from TAPE or do a second
level install using the DVDPRIME you can not install it
The DVD/CD needs the SYSG console which is not supported by Hercules.
Doug
Behalf Of Josef Sipek
Sent: Thursday, June 21, 2007 5:32 PM
Subject: Re: [hercules-390] z/VM on Hercules

Post by herculessearch
Can I run z/VM on top of a Hercules installation?

Yes and no. As far as I know, there's no _technical_ problem, and hercules
can handle it just fine, but you need IBM to let you run z/VM - licensing
issues.
Josef "Jeff" Sipek.
--
The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all progress
depends on the unreasonable man.
- George Bernard Shaw
[Non-text portions of this message have been removed]

Daniel Gaeta
Consultor Mainframe
ddsoft systems
ddsoftsystems-/1Rj0zDCqbBfyO9Q7EP/***@public.gmane.org

[William D Carroll]

Yes, I had a typo in my email I see

"or do a second level install using the DVDPRIME you can not install it"

Should read

"or do a second level install using the DVDPRIME you can install it"









From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of ddsoftsystems
Sent: Friday, June 22, 2007 7:44 AM
To: hercules-390
Subject: RE: [hercules-390] z/VM on Hercules



It's possible install from DVD/CD as a VM second level.
After this you can use to as first level.
I used a z/VM 4.4 to install a z/VM 5.1 from DVD. But you need to use FBA
disks and the correct device model, in this case is 9346.

Post by William D Carroll
Like was stated, Technically it will run but proper licensing is required.
Problem is the installation, unless you install from TAPE or do a second
level install using the DVDPRIME you can not install it
The DVD/CD needs the SYSG console which is not supported by Hercules.
Doug

[Non-text portions of this message have been removed]

[ronx_ronkonkoma]

What about SIE, Hercules can't be implementing that???

Post by herculessearch
Can I run z/VM on top of a Hercules installation?
Thanks in advance for your help.

[Ivan Warren]

Post by ronx_ronkonkoma
What about SIE, Hercules can't be implementing that???

SIE is implemented..

--Ivan

[Jay Maynard]

Post by ronx_ronkonkoma
What about SIE, Hercules can't be implementing that???

Can and does. z/VM ir reporeted to run well under Hercules.

[Adam Thornton]

Post by Jay Maynard

Post by ronx_ronkonkoma
What about SIE, Hercules can't be implementing that???

Can and does. z/VM ir reporeted to run well under Hercules.

I can report that it *runs* under Hercules under Linux/390 under z/
VM. Specifically, 64-bit z/VM 4.4 runs on Hercules on 31-bit Linux
on 31-bit z/VM 4.4.

Adam

[Non-text portions of this message have been removed]

[Raphaël Jacquot]

Post by Adam Thornton
I can report that it *runs* under Hercules under Linux/390 under z/
VM. Specifically, 64-bit z/VM 4.4 runs on Hercules on 31-bit Linux
on 31-bit z/VM 4.4.

sounds like this stack would be horribly slow :D

[Adam Thornton]

Post by Raphaël Jacquot

Post by Adam Thornton
I can report that it *runs* under Hercules under Linux/390 under z/
VM. Specifically, 64-bit z/VM 4.4 runs on Hercules on 31-bit Linux
on 31-bit z/VM 4.4.

sounds like this stack would be horribly slow :D

Affirmative.

Adam

[Non-text portions of this message have been removed]

[William D Carroll]

If you want horrible slow, try loading Debian Linux on an iPaq 5450, Compile
Hercules for an ARM Processor, load it up and fire up MVS on it

A blazing ~0.6MIPS and soooommmmeeee time later, get a Logon screen and it
works. ;-)



But it was VERY, VERY COOL!



Doug





From: hercules-390-***@public.gmane.org [mailto:hercules-390-***@public.gmane.org] On
Behalf Of Adam Thornton
Sent: Tuesday, June 26, 2007 8:27 AM
To: hercules-390-***@public.gmane.org
Subject: Re: [hercules-390] Re: z/VM on Hercules

Post by Raphaël Jacquot

Post by Adam Thornton
I can report that it *runs* under Hercules under Linux/390 under z/
VM. Specifically, 64-bit z/VM 4.4 runs on Hercules on 31-bit Linux
on 31-bit z/VM 4.4.

sounds like this stack would be horribly slow :D

Affirmative.

Adam

[Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]

[Greg Price]

Post by William D Carroll
A blazing ~0.6MIPS and soooommmmeeee time later, get a Logon screen and it
works. ;-)

0.6 MIPS emulated? Still better than the 0.4 MIPS actual real box I worked on.
And I bet you had more memory. The circa 1981 mainframe had 3MB.

(The vendor tendered 2MB, but had to give us 3MB so that the OS would
actully IPL.)

[Paul Raulerson]

And anyone wanting to know how it does it - can look at the sei.c source
code module. -Paul

Post by Jay Maynard

Post by ronx_ronkonkoma
What about SIE, Hercules can't be implementing that???

Can and does. z/VM ir reporeted to run well under Hercules.
--
Jay Maynard, K5ZC http://www.conmicro.com
http://jmaynard.livejournal.com http://www.tronguy.net
http://www.hercules-390.org (Yes, that's me!)
Buy Hercules stuff at http://www.cafepress.com/hercules-390

[Non-text portions of this message have been removed]